Project Summary Of the diverse features of the mammalian motor repertoire, skilled limb movements have become some of the more impressive and indispensable ways of interacting with the environment. The susceptibility of these movements to neurodegenerative disease and injury underscores the need for a better understanding of how neural circuits orchestrate these dexterous behaviors. This goal demands sophisticated experimental scrutiny at both neural and behavioral levels, and while the emergence of genetic tools for monitoring and manipulating neural circuits in mice has been transformative, the development of motor behavioral assays has not kept pace. Typically, a single behavioral test is applied to the question at hand, precluding a more comprehensive description of why relevant neural circuits have evolved their particular anatomical and functional attributes. Moreover, in the study of limb motor control, primate-inspired behavioral paradigms tend to be applied to the mouse by default, risking neglect of a more complete and naturalistic account of motor circuit function. Ethology emphasizes an unbiased study of behavior under natural conditions. This proposal describes an ethological approach for the quantification and categorization of skilled limb movements in mice, enabling a depth and breadth of behavioral analysis that more closely aligns with the complexity of the underlying neural circuits. Several complementary approaches for unbiased behavioral quantification of mouse limb movements in enriched environments will be developed in parallel: a set of optical techniques enabling automated three- dimensional reconstruction of limb and digit posture and trajectory; and an approach that leverages advances in the miniaturization of motion sensors for camera-free limb tracking that can be integrated seamlessly with neural and electromyography (EMG) recordings. Through iterative refinement, this multifaceted strategy should highlight the strengths and mitigate potential drawbacks of each tracking method, providing a suite of complementary quantitative tools. Kinematic, kinetic, EMG and neural data collected across diverse behavioral contexts will be used to guide machine learning-based classification of natural structure in limb movements and in their underlying neural circuits. As a proving ground for how these naturalistic behavioral analyses can be integrated with the genetic dissection of neural circuit function, a set of molecularly defined motor circuits will be probed using novel optogenetic tools that permit selective inhibition at defined axon collateral terminals. This combination of projection specific genetic perturbation and ethologically driven behavioral analyses will provide a powerful lens through which to view the fine-grained functional organization of mammalian motor circuits. More generally, this merging of molecular and systems neuroscience approaches will offer a novel way to explore and compare fine motor control across species, equipping the field with a more comprehensive and standardized approach to study skilled behavior, and helping lay the foundation for better diagnosis and treatment of behavioral deficits associated with neural circuit dysfunction.